Abstract:It is widely acknowledged that urban trees and forests can significantly improve air quality by removing airborne particulate matter (PM) and thus safeguard the health of urban residents. However, it is very difficult to accurately assess the ability of tree species to capture PMs from the air as such processes are governed by several biotic and abiotic factors. Here, we present a first ever study that explores the effects of rainfall on the detention of PMs by leaves in five commonly used urban greening species in northern China, including Ginkgo biloba, Acer mono Maxim, Euonymus japonicus, Koelreuteria paniculata, and Eucommia ulmoides. Our specific objectives were to:1) quantify how much PMs detained on the leaves of five different tree species can be washed off by natural and simulated rainfall events, 2) compare the difference in PM removal between tree species under different amounts of rainfall, and 3) analyze the factors that might contribute to the differences in PM removal under different rainfall conditions. Leaves were uniformly collected from top, middle, and bottom layers of the tree canopy at four cardinal directions from the sampling trees after several minutes of rainfall and 12 days after rainfall. Each sampled tree leaf was sealed in a plastic bag immediately after collection. To measure the quantity of PMs trapped on leaves, the samples were washed and water used for washing was collected. Thereafter, 10% of the water was dried to determine the weight of total suspended particulates (TSP), while the remaining 90% of the water was passed through a 10μm mesh filter and then through a 2.5μm mesh filter. These filters were then dried to determine the weight of PM > 10μm (PM > 10) and PMs that were between 2.5 and 10μm (PM10-2.5). Then, 20% of the filtered water was dried to determine the weight of PMs that were < 2.5μm (PM2.5). Once the quantity of airborne particulate matter held per unit leaf area was determined for each of the study species, leaves were sampled after exposure to the following rainfall conditions:light rain (14mm), moderate rain (29mm), 30mm/h indoor simulated rain (equal to 15mm), and 12 days after rain (to determine the saturated dust held in summer). The result indicated that the amount of PM washed off from the leaves by rainfall varied greatly with the size of the particulates and the rainfall condition (including the intensity and amount of precipitation). For all species except Ginkgo biloba, PM10 could be washed off even under light natural rainfall condition and the amount and intensity of rainfall were critical factors in controlling PM removal. Overall, 50% and 90% of the TSP were removed by moderate rainfall and simulated torrential rainfall, respectively, due to higher intensity of rainfall. One major factor that contributes to the above observations is the surface structure of leaves:smooth surfaces, like leathery leaves, have better ability to trap PMs, which also wash-off more easily. Among the five species, Euonymus japonicus recorded the best overall absorption of particulates (washed TSP:(775.06±33.99)mg/m2, light rain), while Ginkgo biloba had an edge on absorbing particulates of small sizes (washed PM2.5:(426.55±40.83.99)mg/m2, moderate rain). We concluded that the species differed in their ability to trap airborne PMs and in the rate of removal of PM from the leaves by a rainfall event, and this was influenced by the differences in the surface structure of leaves of the species, rainfall amount and intensity, as well as the particulate size.